Deuterium retention in tungsten irradiated by heavy … M. Sakamoto et al . 3rd RCM of CRP on...

Deuterium retention in tungsten irradiated by heavy ions

M. Sakamoto, H. Tanaka, S. Ino, H. Watanabe1, M. Tokitani2

3rd RCM of CRP on Irradiated TungstenIAEA Headquarters, Vienna, Austria

27-30 June 2017

Plasma Research Center, University of Tsukuba, Tsukuba, Japan1 RIAM, Kyushu University, Kasuga, Fukuoka, Japan2 National Institute for Fusion Science, Toki, Gifu, Japan

Contents

M. Sakamoto et al . 3rd RCM of CRP on Irradiated Tungsten, IAEA, Vienna, 28 June 2017 2

Main objective of this study Experimental set up Tungsten sample (TEM observation) Effect of heavy ion irradiation on retention in W

Damage level dependence Flux dependence Sample temperature dependence Simulation results

Summary

Main objective of this studyTo understand the effect of neutron and surrogateirradiation upon microstructure of tungsten and

To understand the effect of the damage of tungsten onhydrogen isotope retention through the observation ofchange in the microstructure

In the last CRP held in Soul, I presented D retention in Wexposed to He plasma in addition to preliminary results ofD retention in W irradiated by heavy ions.

Experimental setup: Cu2+ Irradiation

M. Sakamoto et al. . 3rd RCM of CRP on Irradiated Tungsten, IAEA, Vienna, 28 June 2017 4

Tandetron Accelerator (Kyushu Univ.)Ion: Cu2+

Energy: 2.4 MeVFlux: up to 5 x 1015 Cu2+/m2 sDamage level: up to 4 dpa (max: > 100 dpa)Sample Temp.: up to 873 KExposure area: 8 mmφ

Estimated depth profile of displace-ment damage in W.

Calculation using the SRIM code with displacement energy of 55 eV.

The peak damage region is ~400 nm.10-3

10-2

10-1

100

101

0 200 400 600 800 1000 1200

dpa

Depth (nm)

4 dpa

0.4 dpa

0.1 dpa

0.01 dpa

Heating stage

Experimental setup: Deuterium Irradiation

APSEDAS (PRC, Univ. of Tsukuba) Plasma is produced by RF (13.56 MHz) wave

power (< 5kW, typical: 800 W steady state).==> No electrode, No impurity such as carbon,

Clean environment Magnetic field: < 0.05 T Water cooled sample stage Plasma diameter: ~50 mm Exposure area: φ 8 mm (uniform exposure)

Water-cooled sample holder

φ 50

W sample

φ 8


Plasma exposures have been performed using a compact PWI simulator APSEDAS.

Examined Materials


Base Materials: 1mm-thick tungsten disc (99.99% pure, A.L.M.T. corp.)

• Annealed at ~ 2000 oC for 1 h (Re-crystallized)• The surface was mechanically polished to a mirror finish.

The grain size is in the range of 10 µm to ~100 µm.

100 µm 10 µm

Damage evolution by 2.4 MeV Cu2+ irradiation

M. Sakamoto et al. 3rd RCM of CRP on Irradiated Tungsten, IAEA, Vienna, 28 June 2017 7

Nano-voids (d<1nm) are observed in 1 dpa case and they formed densely.

Most of interstitial loops (IL) is considered to be nucleated by cascade collisions,since the density of ILs was two order of magnitude higher than the estimated valueusing a rate theory assuming that the two free interstitials act as nuclei for ILs..

Each IL can not grow larger individually but additional loops were formed in thevicinity of pre-existing dislocation loops and dislocations and aligned to coalescewith each other.

1.0 dpa 1.0 dpa

Irradiation at room temp.

Experimental Procedure


1mm-thick tungsten discs (99.99% pure, A.L.M.T. corp.)

Annealed at ~2000 oC for 1 h (Re-crystallized)

D plasma irradiation

2.4 MeV Cu2+ irradiation

TDS measurement

Electron density (m-3) 2.7 x 1017

Electron temp. (eV) 10

Space potential (V) 30

Flux (m-2 s-1) 3.7 x 1021

Fluence (m-2) 2.0 x 1025

Surface temp. (K) 480

(1) dpa dependence: 0.01 ~ 4 dpa

(2) flux dependence: 1 ~ 5 x 1015 Cu2+/m2 s

(3) sample temperature: RT, 500 K, 873 K

Damage level dependence of TDS Spectrum

M. Sakamoto et al. 2nd RCM of CRP on Irradiated Tungsten, SNU, Korea 8 Sep. 2015 9

There is three main peaks in thespectra: ~620 K, ~760 K and ~860K.

Temperatures at 1st and 2nd peaks arethe same as those of non-irradiated W.

The 3rd peak newly appears due to 2.4MeV Cu2+ irradiation and is consideredto be caused by vacancy clusters andvoids.

Retention in the damaged W increaseswith the damage level but it saturatesaround 0.4 dpa, suggesting that newlyintroduced defects may be cancelled byalready existing vacancies and voidswith high density.

0

1

2

3

4

5

6

7

0 dpa (x10)0.1 dpa0.4 dpa4 dpa

300 400 500 600 700 800 900 1000

Des

orpt

ion

rate

(1018

D/m

2 s)

Temperature (K)

dpa 0 dpa 0.1 dpa 0.4 dpa 4 dpa

Peak 1 613 600 624 634

Peak 2 759 761 770 765

Peak 3 861 858 863

5 x 1015 Cu2+/m2 s

Flux dependence of TDS Spectrum


Comparison between high and low fluxesof Cu2+ ions indicates that D retention inW irradiated with low flux Cu2+ ionsbecomes 3.5 times lower than that forhigh flux irradiation.

The desorption spectrum of the low fluxirradiation also consists of three stages ofdesorption.

The temperatures at the three peaks forthe low flux case became lower thanthose of the high flux irradiation.

0

1

2

3

4

5

6

7

5 x 1015 m-2 s-1

1 x 1015 m-2 s-1

300 400 500 600 700 800 900 1000

Des

orpt

ion

rate

(1018

D m

-2 s

-1)

Temperature (K)

High flux

Low flux

0.4 dpa

Comparison between high and low fluxes of Cu2+ ions

Flux 1 x 1015 m-2s-1 5 x 1015 m-2s-1

Peak 1 553 K 624 K

Peak 2 724 K 770 K

Peak 3 825 K 858 K

Comparison of dpa dependence between high and low irradiation fluxes


In the case of high flux irradiation, D retention increases with the damage level butit saturates around 0.4 dpa.

This suggests that newly introduced defects are cancelled by already existingvacancies and voids with high density.

In the case of low flux irradiation, D retention increases with the damage level up to2 dpa and no saturation is observed.

It is found that defect formation due to heavy ion irradiation depends on not only adamage level but also flux of the high energy ions.

Simulation of TDS spectra using an HIDT code


0

1

2

3

4

300 400 500 600 700 800 900 1000

Des

orpt

ion

rate

(1018

D/m

2 s)

Temperature (K)

The HIDT (Hydrogen Isotope Diffusion and Trapping) code has been developed in Shizuoka Univ.

(ref: Y. Oya et al., JNM 461 (2015) 336.) Input parameters:

D = 2.9 x 10-7 m2/s (Frauenfelder)Kr = 3.0 x 10-25 m4/s (Pick)

Time evolution of sample temperature and D ion flux for the simulation was the same as those of the experiments.

A depth profile of defect concentration is similar figure to the dpa profile calculated by the SRIM code.

0

200

400

600

800

1000

0

1 1021

2 1021

3 1021

4 1021

5 1021

0 5000 10000 15000 20000

T(K)

Imp(atom/m2/s)

Temp

eratu

re (K

) Imp(atom/m2/s)

t(s)

Plasma

Sample temperature

D ion flux

TDSMove to TDS system

Exp. Data 0.1 dpa (high flux)

Sample temperature & D ion flux vs time

0

0.002

0.004

0.006

0.008

0.01

10-11 10-10 10-9 10-8 10-7 10-6 10-5 10-4 0.001

T1 (0.83 eV)T2 (1.22 eV)T3 (1.46 eV)

Depth (m)

Con

cent

ratio

n (a

tom

-1)

Defect concentration of each binding energy



0

1

2

3

4

300 400 500 600 700 800 900 1000

Simulation0.83 eV1.22 eV1.46 eVExperiment

Des

orpt

ion

rate

(1018

D/m

2 s)

Temperature (K)






0

200

400

600

800

1000

0

1 1021

2 1021

3 1021

4 1021

5 1021

0 5000 10000 15000 20000

T(K)

Imp(atom/m2/s)

Temp

eratu

re (K

) Imp(atom/m2/s)

t(s)

Plasma

Sample temperature

D ion flux

TDSMove to TDS system

Sample temperature & D ion flux vs time

0

0.002

0.004

0.006

0.008

0.01

10-11 10-10 10-9 10-8 10-7 10-6 10-5 10-4 0.001

T1 (0.83 eV)T2 (1.22 eV)T3 (1.46 eV)

Depth (m)

Con

cent

ratio

n (a

tom

-1)

Defect concentration of each binding energy

Well reproduced



0

1

2

3

4

300 400 500 600 700 800 900 1000

Simulation0.83 eV1.22 eV1.46 eVExperiment

Des

orpt

ion

rate

(1018

D/m

2 s)

Temperature (K)






Well reproduced

PeakBindingenergy

Corresponding defects

1st 0.83 eV dislocation loops & grain boundary

2nd 1.22 eV vacancy

3rd 1.46 eV vacancy cluster & void



High ion flux

Low ion flux

0

2

4

6

8

0 1 2 3 4

5 x 1015m-2s-1

1 x 1015m-2s-1

Ret

entio

n (1

020 D

/m2 )

Desorption stage

72%33%

84%

D retention in the 2nd stage (i.e. vacancy) significantly decreased with decrease in ion flux (i.e. dpa rate), while D retention in 3rd

stage decreased not so much.

Vacancy

Dislocation Viod

High dpa rate: vacancies may be remained.Low dpa rate: voids may be produced.

Speculation

Dependence of D retention on sample temperature druing Cu2+ ion irradiation


Cu2+ irradiation was carried out at the sampletemperature of RT, 500 K and 873 K.

Desorption rate decreased as a whole with thesample temperature during the irradiation.

For the sample temperature of 500K, 873 K, Dretention decreased by 60 % and 80 %,respectively.

0.0

0.5

1.0

1.5

2.0

300 400 500 600 700 800 900

D re

tent

ion

(1021

D/m

2 )

Sample temperature (K)

0

1

2

3

4

5

6

7

300 400 500 600 700 800 900 1000

4dpa/RT4dpa/500K4dpa/873K

Temperature (K)

Des

orpt

ion

rate

(1018

D/m

2 s) Cu2+ ion flux: 5×1015 m-2s-1

Two possible candidates1. Interstitial and vacancy may annihilate each

other with increase in the sample temperature.

2. At the sample temperature of 873 K,Interstitials may escape to grain boundary andremained vacancies coalesce to become void.

QuestionWhich is dominant for D retention, large number of vacancies or smaller number but bigger voids?

Summary


Deuterium retention in tungsten (W) samples irradiated by 2.4 MeVCu2+ ions hasbeen investigated to study effects of the damage on hydrogen isotope retention in W.

Most of the dislocation loops (ILs) were nucleated by cascade collisions. Additional loops were formed in the vicinity of pre-existing dislocation loops

and dislocations and aligned to coalesce with each other. Nano-voids (d<1nm) are observed in 1 dpa case and they formed densely.

TEM observation

There exists three desorption peaks in the TDS spectra . The desorptionpeak around 850 K newly appeared due to Cu2+ ions irradiation. This peakmust be related to voids.

Flux and damage level dependences

TDS spectra

When the Cu2+ ion flux decreased by 5 times, the D retention decreased by3.5 times, indicating clear flux dependence.

In the case of higher ion flux, D retention increases with the damage levelbut it saturates around 0.4 dpa, suggesting that newly introduced defectsmay be cancelled by already existing vacancies and voids with high density.

Summary (continued)


On the other hand, In the case of low flux irradiation, D retention increaseswith the damage level up to 2 dpa and no saturation is observed.

Flux and damage level dependences

Sample temperature dependence Desorption rate decreased as a whole with the sample temperature during

the irradiation.

Simulation using the HIDT code

TDS spectrum was reproduced by using the HIDT code using experimentaldata. The binding energies of three desorption peaks are 0.83 eV, 1.22 eVand 1.46 eV.

It is necessary to know effects of the number of defects and their size on D retention. It is planed to measure TEM using samples which have irradiated at different sample temperature.

High dpa rate: vacancies may be remained.Low dpa rate: voids may be produced.

Speculation

Deuterium retention in tungsten irradiated by heavy … M. Sakamoto et al . 3rd RCM of CRP on...

Documents

Transcript of Deuterium retention in tungsten irradiated by heavy … M. Sakamoto et al . 3rd RCM of CRP on...